The Bud Toaster - (currently: Model 14, version 3)

[Hippie Dickie]

Just place the normal operational connections on jumpers

can't do that for the production pcb, since i need to program it after the PIC is soldered on to the pcb.

my first theory is that MAX pin 7 is holding the PIC's MCLR (pin 4) low so the PICKit1 can't raise it to initiate programming. Maybe a 1k resistor will allow the pin to float enough for the PICKit1 to raise the level.

 

[Hippie Dickie]

i think i found my answer ...

For example if you put an LED on RB6 (PGD) that draws 20mA when on. The output voltage (when the output is on) will be pulled so low that the pic ICSP programmer will not be able to read back the desired voltage i.e. it will give a verification failure.

... because verification failure is exactly the error i am getting. And i do have a LED on that pin.

 

[digglover]

What a gorgeous device!
I'm looking forward to seeing this develop even more! Great job Hippie!

 

[Hippie Dickie]

Thank-you.

Here is the new construction. i switched the bottom-side layout which moves the MOSFET to the opposite end from the MAX6675. This allows better access to the through-holes for the k-type thermocouple.

Anyway, i am so utter confident that this is the "final" production prototype that i thought i'd make three Bud Toasters at once:

(1) Here is the trace mask and blank copper-clad pcb:

(2) Here is the top layer ironed onto the pcb (and drilled):

(3) and bottom layer:

(4) etched and cut apart:

(5) cherry wood discs:

(6) family photo - mother and zygotes:

 

[sneezyjesus]

You have no idea the faith in creative abilty you have restored in me thanks to this post.

I mean, seriously. I'm used to an internet full of multilayered-crap, then I stumble across this site and sebsequently a post about making a vape from scratch, and using reasonably intricate and clever electronics and engineering at that...

Keep up the awesome.

 

[Hippie Dickie]

Thank-you sneezy ... i love surfing the edge of the creativity flow ... this is my first time in hardware, which is enough like software these days it can be considered to be "physical software".

The first son of the Mother is emerging ...

front side, populated:

back side:

The sharp eyed reader will immediately recognize that i swapped polarity of the input lines so that i could relocation the MOSFET away from the MAX6675 on the other side of the pcb.

Also, i installed 6 contact pins (those gold things above the 3 buttons) taken from a spare 8-pin DIP socket so i can just plug in the 3 LEDs instead of soldering them in.

Progress has been slow ... busy week it was.

 

[VivaSativa]

This looks very impressive, and I love your open source approach
Any chance of seeing a video of this vaporizer in action? The whole process of loading the weed, heating it up and getting a lungful of vapor?

 

[Hippie Dickie]

Yeah, i should do a video ... after i get the duty cycle algorithm tuned up a bit ... i want that green LED to come on and stay on.

On another note, i solved the problem i was having when i tried to do In-Circuit-Serial-Programming. It turns out that pin 7 of the MAX6675, the data output pin, was holding pin 4 on the PIC (MCLR) low so the programmer can't raise the logic level to initiate programming. So i put a 2k ohm resistor on the line between the PIC and MAX and now the programmer works correctly. This is a good thing to do anyway, since the MAX doesn't want to sink or source more than 1.6 mamp on that line. A 2k ohm resistor is the exactly right value to limit the current and it allows the pull down of the MAX to be by-passed by the programmer.

Hippie is a happy boy!

(i really love how these problems solve themselves)

 

[ComplexMessiah]

Hey Hippie Dickie,

First time posting over here at FC, I recently embarked on trying to build my own vape and just devoured this whole thread when I came across it last night. Thanks for sharing all your trials and evolutions. Your vape seems great and I'd def love to get one once it hits production.

While reading through all the design revisions something struck me when you to returned to the current cylindrical body. (Sorry to chime in so late since you seem to finally have a pre-production model and I know you've gone through ALOt of 'em) but my inner Industrial Design/Ergonomics guy thinks the vape could be really handy if the power connectors were on the bottom. The batteries (or AC adapter) could be housed in a cylinder of equal diameter to the main housing which the vape could dock on top of. The control buttons would be moved 90 degrees to the side (next to the LEDs in the current design)

Seems like it might have a bit more ease of use/portability...you don't have to turn the unit over to adjust temps, and you can lift the unit and power source together in one hand as a stable connected device without any trailing wires...while still having easy detachability for recharging/swapping in new batteries and switching to AC or DC.

Anyway, thanks again, look forward to keeping updated on the Bud Toaster's journey!

 

[Hippie Dickie]

CM - thanks for your comments ... i'll digest them for a while, but i have some initial comments:

re: AC adapter - it has to be able to deliver 12 amps at 5v (or more) DC. The 3.3v voltage regulator has a 1.7v dropout, so the voltage has to be >5v. There is no AC adapter (that i have found) that can deliver 12 amps in a package as small as the two cell battery pack.

For my personal use, i will never go back to AC - i love being able to grab the vape and go anywhere (with a bandoleer of batteries strapped across my chest, of course).

i still need to try plugging it into a car battery and a DeWalt drill power cell. i can imagine a cannabis coffee shop having a row of Bud Toasters plugging into a Peterbuilt battery.

Each cell is 1.065" in diameter by 2.625" long. The vape body is 1.75" dia by 3" long. Vape body could be bigger diameter to get the two batteries side by side, but that would be more cumbersome to hold in the hand. If the batteries are stacked under the vape, the whole unit would be 6" (or so) tall, which i would sure as shit knock over - as i have other design versions that were taller than 3" (stupid stoner move, for sure). And the unit would be too heavy to pick up and toke -- based on experience with an earlier test unit that included the batteries.

The Bud Toaster (cylinder version) just feels too good in the hand to not be able to pick it up to toke -- but maybe that's just me.

i tried a "control panel" of buttons on the side, and it is too easy to accidently hit one of the buttons when holding the vape body. Maybe they could be recessed ... but then the control panel needs to be fastened or grabbed. i tried working through a number of configurations and couldn't find one i liked.

re: turn the unit over ... this may be a fatal flaw in the button placement, but i'm trying to think through to past the prototype stage, when everything is working properly. Except for the initial power on, it is not necessary to even mess with the buttons -- it remembers the last temp setting and automatically returns to it. i find (from extensive, exhaustive, daily (sometimes hourly) testing) that i rarely mess with the temp once it is at a good setting.

However, the first time i stonerly flip the vape over to press a button and drop the 400F vial into my lap, i will relocation the buttons.

Maybe the Bud Toaster should be controlled via an iPhone app -- but i think i'd need a PIC with more output pins to control the WiFi. Maybe on the high(er) end version -- i can see lots of variations once the Bud Toaster goes into production -- got to keep ahead of the Chinese, eh?

In an earlier prototype i did have a connector on the vape, but it came out the side (i've always wanted to put the vape down on the desk on its bottom), and then the connector on the cable is cumbersome when holding the vape.

One of the design limiters is the connector. i swear by these Deans R/C plugs - they are the only thing i've found that can reliably carry high current loads. Super reliable, unbelieveably reliable - with maximum current flow. The radio control crowd KNOWS massive current flow.

That being said, i really hate having the cable, but that's the way it is for now. Better batteries will change this in a few years.

And, again, i appreciate your comments. i get locked into a viewpoint and only occasionally can get a fresh perspective.

 

[Hippie Dickie]

Just had to take a day off work today ... so i spent the day in the vape shop ... and here it is:

First son of the Mother (production prototype):

Despite the thread title, this is actually Model 13, Version 5-4d. It still needs the shell of stainless steel, leather and ceramic tape to be complete.

i just checked and i can program the PIC which is soldered onto the printed circuit board without any hassle or problems, which is very exciting because i now have an easy to access vape to complete the development of the duty cycle algorithm -- specifically, tuning the PID of the algorithm to eliminate the 5F swing i've been experiencing by not having the "I" part of the PID implemented.

i'm going to have to retrofit Mom with a 2k ohm resistor to isolate the MAX6675 from the PIC to allow in-circuit programming like the Son enjoys.

Also, while i was embracing some demons, i got out the glass drilling rig and finally (FINALLY!) drilled a larger whip tube. This pix shows both the standard whip (the short one on the bottom) which is 1/2" dia by 4" long, and the new longer whip which is 5/8" dia by 6" long. It is the size and feel of a large cigar:

i think i prefer the larger tube -- i can see more vapor, it feels like a fuller hit, it seems to pull more vapor for a smaller load -- but these are still very preliminary findings.

First vapors at 11:00PM.

 

[vtac]

It's like reading the diary of some 19th century inventor or scientist. Some day people will understand how awesome this is. Can't wait.

 

[stonemonkey55]

any videos of this thing up and running?

 

[Hippie Dickie]

not yet ... i'm making a few mods in the software and i want the video to accurately reflect what the Bud Toaster is capable of.

i've been cruising in slacker-stoner mode for the last week and plan to continue for just a few more days ... although i did take a 1500 mile road trip over the last weekend -- and worried about my kitties the whole time i was away from home.

And running my business. Big changes coming this year and that's got me a bit engaged in things non-vapor ... except for tokin' on the Bud Toaster, of course -- i'm always running that baby.

 

[zombonaut]

This is the coolest thing ever. I have no skills but I would really love to buy one of these beauties!

 

[Progress]

Hello Hippie!
This is definitely one impressive project you have going on.
I know that I would definitely be interested in becoming an early adopter once you get all of the bugs worked out (take all of the time you need to perfect it, as it will likely benefit all of those who use it, including yourself in the long run :wink: ).
Have you been flame polishing the holes in the glass to allow for easy cleaning (and does/could the glass 'oven' protrude a bit to protect the surrounding wood from the possibility of contamination through vapor leakage upon the removal of the stem). Just some outsider feedback (toke it with a grain of salt) Keep up the good work (and please do continue to keep us posted)

 

[Hippie Dickie]

Thanks, zombonaut.

Progress - thank-you for your comments and interest in this project.

i use a toothpick to clean the holes, and so far i haven't tried flame polishing -- i'm not sure it is necessary -- i wonder if a laser engraver can drill a hole in glass, that would be so fucking cool -- what a time saver that would be!

i keep inhaling as i pull the stem away from the oven tube to clear the vapor in the stem. i'm not seeing any leakage. i like to keep the lip of the oven tube slightly recessed so i can't accidently touch the 3xxF glass ... it is totally safe to hold 440F in the palm of my hand. i get a kick out of that.

i do try to fire anneal the lip of the oven tube after i cut it (quick check of Wiki tells me i'm not doing it long enough). That is currently the weakest link in the fabrication process -- that is, once i can outsource the pcb fabrication.

Keeping the glassware clean is just not a problem. i'm just a bit OC about clean vape tubes, and borosilicate glass is a dream to keep spotlessly clean. i love vaping with a sparkly clean tube.

 

[Bubar]

Have you thought about mounting the buttons in the bottom wood disc opposite the LED? I think that would make sense because the user could reach around and mess with the buttons while watching the led.

Anyway, I got back from vacation and I was itching to get the vape working right. I've been without a properly working set up for a month. I rigged up a little ghetto box with a dedicated molex connectors rather than alligator clips. Sorry bout the cell phone pics.

Worked pretty good on a 4v, 5A supply. I took it up to 12v and the transistor melted up and was smoking. It did heat up almost instantly tho. Tomorrow I'll put a new transistor in and rewire the molex to only use the 5v. Got another roll of nichrome coming in the mail too.

 

[B.]

It's like reading the diary of some 19th century inventor or scientist. Some day people will understand how awesome this is. Can't wait.

My thoughts exactly. Someday. I feel the same way about vaping in general. I just cannot understand why more people don't vape. For cigs too.

 

[Hippie Dickie]

Bubar - i feel your pain ... i get quite frantic when my only working vape craps out. i feel quite secure these days with two working vapes ... well, i had three for a while, but i had to scavenge one of them for body parts for the second one which i'm using as my main vape.

Why did the MOSFET melt? How fast was the temp rising on 12v? Shouldn't the PWM control the current? Was that running my code? (easier to ask questions than run a test, eh?)

Now i'm struggling with the pcb design software --- why can't i just say "12F683 in SOIC" and have the footprint appear on the grid, ready to route?

 

[Bubar]

In most pcb design software you should be able to select a generic 8-pin soic to lay down the pads.

It was running my software. The mosfet melted cuz i overloaded it with approx.18-20A. It heated up to 160C within 5-10s or however long it took me to notice the smoke. Future versions of my pcb will use ground plane on the back and some extra copper on the front for addition heat sinking. I will probably try to stay within 12-15A in the future. I don't remember what the ratings of my mosfet were.

I'm having trouble with the mounting of the thermocouple. What is your method? I tried putting in a hole in the pcb, but i made it to small. Currently I put some 30awg wire through the hole and then wrap it around the thermocouple tips, then solder around the wrap.

 

[vap999]

Hippie:

Rather than drilling holes in glass test tubes, have you ever considered using fritted (porous, sponge-like) glass disks, i.e., using these as a screen, much as you might use a metal screen to hold the herbal material? These are commonly used in laboratories for filtration. The disks could be dropped in and out, cleaned and might even be cheap enough to be disposable. For example, see http://www.wilmad-labglass.com/group/1213

 

[Bubar]

I've also done slits, which are much easier than holes and with a dremel and abrasive wheel.

 

[Hippie Dickie]

Bubar - whew ... glad it wasn't my code, which i expect to handle the higher voltage situation. The MOSFET i use (Fairchild FDB6030BL) handles up to 30 volts and up to 40 amps. My coil should only draw 24 amps at 12 volts.

The K-type thermcouples i get are bare wire, 0.010" dia, and have 12" leads, which i cut down to 3". The leads insert from the backside of the pcb and stick up through the surface-mount side (top side). i bend each end into a loop and bury it in solder on the top side.

i insulate the leads with Zeus PTFE (Teflon) Sublite Wall Tubing 28 AWG .013" ID x .0030" before soldering into the pcb.

Here's the backside view, where the k-type probe leads go through the pcb:

Here's where the probe leads are soldered on the topside:

i stuff the probe tip, along with the probe for a digital kitchen thermometer into a piece of 16ga teflon tubing.

Then i stuff that piece of tubing between the loops in the middle of the coil.

This works amazingly well. The probes stay in place and the teflon is against the glass oven tube.

i can't find a SOIC footprint in the provided libraries that matches any of the dimensions on the MAX6675 or 12F683 datasheet (SOIC Narrow Body). So, finally, this afternoon, i designed my own footprints. One more dragon slayed. i should have the pcb routed soon.

 

[Hippie Dickie]

vap999 - i briefly considered fritted glass - i'll explore that link, thanks.

Actually, drilling the glass is not that hard with the right bit and jig setup.

i did use slits before i got the diamond drill bits, but the slits i had were way too large to be useful as a screen.